Science Service System

Summary of Proposal MTH0215

TitleKnowledge based Image Information Mining (KIM) for TerraSAR - X data exploration: Development and validation of scenarios and applications of the KIM module of the TerraSAR-X Payload Ground Segment
Investigator Datcu, Mihai - German Aerospace Center (DLR), Remote Sensing Technology Institute (IMF)
Team Members
Mr. D'Elia, Sergio - ESA, ESRIN
Dr. Seidel, Klaus - ETH, BiWi
Dipl. -Ing. Valente, Achile - ACS, EO
Dipl. Zelli, Carlo - ACS, EO
Dipl. -Ing. Schwarz, Gottfried - DLR, IMF
Dr. Quartulli, Marco - ACS, EO
Dr. Iapaolo, Michele - ESA, ESRIN
Dipl. -Ing. de Miguel, Amaia - DLR, IMF
Dipl.Ing. PhD studen Espinoza, Daniela - DLR, IMF
Dipl.Ing. PhD studen Cerra, Daniele - DLR, IMF
Dipl.Ing. PhD studen Soccorsi, Matteo - DLR, IMF
Dipl.Ing. PhD studen Chaabouni, Houda - DLR, IMF
Rosati, Claudio - ACS, EO
SummaryTerraSAR-X (TS-X) will provide the scientific community with spaceborne X-band SAR data which will strongly reinforce to potential of SAR. TerraSAR -X will acquire daily a volume of up to 100 GB of high complexity, multi-mode SAR images, i.e. SpotLight, StriMap, ScanSAR, dual-, or quad-polarization, different look angles . The high resolution and multiple resolution of the instrument, 1 m, 3m and 10m, opens perspectives for new applications, not possible for lower resolution sensors (20-30m) in the past. Mainly the 1m and 3m modes are expected to support a broad range, and stimulate new applications mainly in relation with human activities which results in scenes with relevant structures and objects at the 1m scale. Thus, among the most interesting scenes are: urban, industrial, rural, etc. In addition the global coverage and the relatively frequent repeat pass will definitely help to acquire extremely relevant data sets. However, the today state of the art methods for SAR information extraction and understanding are not yet sufficiently developed to cope with the dual curse of complexity: high resolution and huge data volumes, also the methods for data quality are still rather laborius, not precise enough and time consuming. To this end, we proposed a novel advanced remote sensing information processing systems Knowledge-driven Information Mining (KIM). KIM is based on Human Centred Concepts (HCC), which implements new features and functions allowing improved feature extraction, search on a semantic level, the availability of collected knowledge, interactive knowledge discovery and new visual user interfaces. We propose a demo of the new technologies for managing, asses quality, exploring, extract information and adding value on a large set of TerraSAR-X products. The proposal also aims at development and demonstration of new SAR data use and novel applications, specific for high resolution SAR and using large data volumes. The research activities regarding this proposal are in relation with: the ESA project PIMS- Partner Information Mining System, and ESA project KEO-Knowledge Centred Earth Observation. Data requirements: 1. SpotLight mode: 25 geo-referenced scenes single polarization un urban/industrial sites 2. StripMap mode: 25 geo-referenced scenes single polarization un urban/industrial sites 3. SpotLight mode: 5 scenes dual polarization un urban/industrial sites The research activities regarding this proposal are in relation with: - ESA project PIMS- Partner Information Mining System - ESA project KEO-Knowledge Centred Earth Observation - DLR program for integration of KIM in the TerraSAR-X Payload Ground segment - The activities of the CNES/ENST/DLR Competence Center for Information Extraction and Understanding of EO data. The planned activities are: - Data processing, for ingestion in the KIM system of all proposed scenes. The ingestion be effectuated on the DLR premises in Oberpfaffenhofen. The original products will remain internal at DLR. - system and technology tests - validation of operation and functionality - opening to the consortium and to the experts and evaluators access to the KIM system via a client operated over internet. - training of the evaluators and experts for KIM operation - coordination and monitoring of the access to SAR data information and knowledge Deliverables: - accessing the TerraSAR data archive - asses quality and information content of TerraSAR products - perform classification, target detection and discover structures in scenes observed - elaborate and propose new TerraSAR data use scenarios and new applications - demonstrate the results

Back to list of proposals

DLR 2004-2016